Spring 2014

Uncertainty in Risk Analysis during Hurricane Katrina

Monika

Radomińska Peter Frederiksen

1 | P a g e

Abstract

This project aims to determine the reason as to why risk analysis in hurricanes is uncertain.

Risk analysis is carried out to determine the proper way of dealing with the risk at hand. In

natural disasters, risks are thoroughly assessed and using appropriate methods, dealt with in a

way which lowers the threat. Hurricane Katrina is infamous for its devastating strength that left

tens of thousands homeless across five states in United States, and close to two thousand

fatalities. Having caused over $125 billion of damages, Katrina became one of the most

expensive natural disasters in near history. The vast damages and fatalities during Katrina’s risk

analysis proved that the process is not perfect. Therefore with the aid of literature and news

articles a detailed timeline of Hurricane Katrina was created to study the events which took

place during the risk analysis.

Finally, although there were several factors which contributed to the devastation, the

uncertainty in forecasting and therefore limited amount of time for decisions to be made is the

main factor responsible for the damages during Katrina.

2 | P a g e

Table of contents

Abstract ............................................................................................................................................................................ 1

Table of Contents ......................................................................................................................................................... 2

Introduction ........................................................................................................................................................................ 3

Problem formulation .................................................................................................................................................. 4

Semester requirements ............................................................................................................................................. 4

Methods ................................................................................................................................................................................ 5

Results ................................................................................................................................................................................... 6

Risk analysis .................................................................................................................................................................. 6

What are hurricanes? ................................................................................................................................................. 8

Monitoring of hurricanes .................................................................................................................................. 10

Weather satellites ............................................................................................................................................ 11

Radars ................................................................................................................................................................. 11

Hurricane hunters .......................................................................................................................................... 11

Weather buoys and ships ............................................................................................................................ 12

Consequences of hurricanes ............................................................................................................................ 12

Hurricane Katrina .......................................................................................................................................................... 13

Discussion ......................................................................................................................................................................... 18

Conclusion ........................................................................................................................................................................ 21

References ........................................................................................................................................................................ 22

3 | P a g e

Introduction

Risk analysis is the overall process of monitoring, analysing and finally either acceptance of risk

or mitigation of the risk. The process allows for the correct assessment as well as reaction to the

threat, whilst prioritizing human life. Risk analysis takes into account the risk’s location,

strength, and hazards which may accompany the primary risk. This permits for further

dissection of the threat and in case of hurricanes, assists in hurricane forecast as well as

mitigation plans. However, as natural disasters are borderline unpredictable, the risk analysis is

that much more difficult, and frequently severe damages still occur, depicting that risk analysis

process is not 100% perfect.

Natural disasters are worldwide occurring phenomenon which can bring forth incredible

damage to fauna, flora, infrastructures, and human life. These devastating catastrophic events,

such as floods, volcanic eruptions, landslides, hurricanes and earthquakes, to list a few, cause

major disruption to anyone and anything in their path. Natural disasters have affected over a

billion people in past two decades, resulted in extensive economic damages, and killed millions

of people (Watson et al., 2007).

Hurricanes are one of the most frequent and destructive natural disasters which occur on planet

Earth (Emanuel at al., 2006). Although usually quickly discovered, they still manage to cause un-

repairable and often quite costly damages, as well as cause several deaths during their lifetime.

Hurricane Katrina was one of the most expensive, deadly and devastating natural disasters to

occur in near history, killing 1833 people across five states in United States of America. Katrina

was the strongest hurricane to occur during the already harsh 2005 Atlantic hurricane season,

and managed to leave tens of thousands of people homeless. Hurricane Katrina’s forceful winds

caused enormous storm surges which flooded the coast along the Gulf of Mexico and left the city

of New Orleans underwater. For this reason precisely, risk analysis is used to identify the risk,

and using value judgements and tolerance criteria, establish the safest mitigation for the threat.

Hurricane Katrina is a perfect case study to examine, as its death toll and vast damages prove

that something in risk analysis must have gone astray. Although the threat was discovered early

on, the immense flooding and incredible winds have still managed to cause massive damage and

close to two thousands fatalities. Therefore, to determine why lives were lost and cities

destroyed, internal and external factors involving risk analysis and its carrying out have to be

thoroughly assessed.

4 | P a g e

Problem formulation

The goal of this 4th semester project is to evaluate the risk analysis and timeline of events during

an event of a natural disaster – a hurricane. The importance of understanding of events which

took place during a hurricane allows for pinpointing the areas of risk analysis or other factors

which have influenced the outcome. Therefore I plan to evaluate the factors which played a role

during risk analysis of an infamous hurricane Katrina, to further understand why damages still

occurred.

Main question:

Why is risk analysis of hurricanes uncertain?

Sub question:

What factors influence the outcome of risk analysis during a hurricane?

Semester requirements

As a fourth semester basic project, I am allowed to select any subject within the natural science

field. I decided to work with analysis of risk assessment as I have worked with it before and find

it relevant to my interests. This time, however, I focus more deeply on the factors, both internal

and external, during a risk analysis process during a hurricane. Natural disasters are also

something I find highly interesting, especially hurricanes, due to experiencing several of these in

my lifetime. I intend on focusing this project on both the scientific and somewhat social aspect

of science, as including roles of people and the effect on their personal and regional economy

will be included, respectively.

5 | P a g e

Methods

The project is intended to be rooted in natural science; however, aspects of social science are

included as well. The nature of the study is qualitative, meaning I am mostly interested in the

details surrounding the event. This will allow me to determine the factors influencing the risk

analysis during a hurricane. Economic aspects will be included as well, due to damages being

often measured in monetary value.

I will be looking into a case study of a hurricane, and breaking down its timeline, to further

understand the events which took place. The criteria for determining which case study to cover

was based on amount of damages and deaths which occurred, therefore proving the aim of risk

analysis was not successful. Furthermore, it is of importance to work with a case study which

has a vast amount of literature based on it, as detailed timeline is required.

The methods used in this project are purely research based; therefore no actual modelling or

experiments are being done. The research on hurricanes, risk analysis and a case study is being

performed by the use of literature from various external sources, as well as online articles,

books and publications. Finally, a United States hurricane Katrina was selected, due to its

enormous data collection, and billions worth of damages. Hurricane Katrina was also

responsible for over a thousand deaths in its lifetime. When looking into hurricane Katrina’s

timeline, I plan to focus solely on the events during the hurricane, rather than looking at the

aftermath. The events which took place during the hurricane are of most importance as it is

then that the essential risk analysis took place when concentrating on actions which led to the

outcome. However, a less detailed look at the actions which occurred during the aftermath is

still required to quantify the losses.

6 | P a g e

Results

Risk analysis

Risk analysis is the process of identification and analysis, followed by either mitigation or

acceptance of said risk. It is used for several purposes, such as risk involved in companies,

individuals or, in this case, natural disasters. Natural disasters, being usually highly destructive

and in many cases- unpredictable, are in need of being handled in a way which lowers the risk

toward people, nature and infrastructures. Hurricanes, being a huge force, require ongoing

monitoring and timely management of the threat. As hurricanes tend to form in different

locations and different magnitudes, the scientists in charge of models are inclined to analyse

and prioritize modes of actions. Furthermore, all of risk analysis’ main priority is human safety,

no matter which type of risk is involved (Ready).

Important part of risk analysis is comparison and data collection from previous hurricanes.

From looking at the past, statistics regarding hurricane hot spots can drafted, therefore

maximizing monitoring in areas prone to heavy storm activity. Looking into this database can

shed light on patters of hurricane movement and pathway, helping scientists in hurricane

forecasting.

Risk analysis deals with identifying and defining the risk at hand, as well as its scope. In Figure 2

a division in risk analysis into two parts can be seen: hazard analysis and consequence analysis.

Danger characterization is a thorough study of the risk, used to determine its limits, strengths,

and overall assessment. Once a risk has been identified, it must be analysed, therefore hazard

analysis deals with understanding whether or not the risk is a viable threat. Hazard analysis

furthers the understanding of the risk, where it’s located and its formation, while also setting

geographical limits, frequency, timeframe and speed of the hazards occurrence. Consequence

analysis, however, deals with identifying the consequences which may accompany the natural

disaster (Hungr et al., 2005).

7 | P a g e

Figure 1. Risk analysis flowchart (Hungr et al., 2005)

8 | P a g e

In risk analysis of hurricanes, defining the location is a very important step, as depending on

location, different scenarios may occur. Consequences in hurricane risk analysis can often be

more damaging and life threatening than the hurricane itself. Along the coast, storm surge and

tide can cause devastating damages, whilst heavy rainfall can cause flooding inland. Winds

themselves are dangerous and can rip trees from the ground, or move cars, jeopardizing

surrounding environment and general public’s safety. Those winds can also aid in formation of

tornadoes on land, or rip currents in the ocean, endangering ships and people aboard them.

Knowing all this allows for predicting several scenarios and constructing event trees or hazard

maps to determine possible results. Accounting for consequences also considers vulnerability of

not only the hurricane but also area which might be also affected by its consequences. Analysis

of probability is the next step in risk analysis, which focuses mainly on probability of occurrence

of the consequence, as well as calculation of possible severity. Using probability and

vulnerability models makes it possible to pinpoint areas prone to most danger, therefore

prioritizing areas at risk.

Last step in risk analysis is risk estimation, which accounts for the information regarding the

longevity, intensity, magnitude and probability of occurrence (Hungr et al., 2005).

What are hurricanes?

Tropical cyclones are called different names depending on their location and strength; however,

as the case study, hurricane Katrina occurred in United States, it is properly classified as a

hurricane. Hurricanes are defined as large, rotating thunderstorms which form over tropical

and subtropical bodies of water (NASA, 2012). They are characterized by spiral storms, with a

low pressure air in the centre of the eye and high speed winds. As shown in Fig. 1, the eye of the

hurricane is centred in the middle of the storm and usually ranges from 30 to 65 km across.

Convection in a hurricane causes bands of vapour filled air to rotate around the centre, forming

an eye wall. Surrounding the eye are the strongest winds, and rotating winds cause updrafts,

therefore moving the air from ocean’s surface toward the top of the storm (Fig. 1). Majority of

the updraft air flows over the storm and back down around the outer edge of the storm. On land,

the eye is the serenest part of the storm, where skies are clear of wind, rain and clouds, over

water, however, the eye creates enormous waves that can reach up to 40 meters in height

(Wolchover, 2011).

9 | P a g e

Figure 2. Cross section of a tropical cyclone (http://www.tcpalm.com/news/2004/may/27/graphic-the-anatomy-of-a-hurricane)

Before becoming an actual

hurricane, the storm must go

through 4 distinct stages: tropical

disturbance, tropical depression,

tropical storm and finally, a

hurricane.

The first stage marks the

beginning of a tropical weather system, identified by slight wind circulation around an

area of low pressure. To become a tropical disturbance, it must uphold its identity for at

least 24 hours and be around 200 to 600km in diameter.

Furthermore, to be classified as a tropical depression, the weather system must reach a

sustainable wind of at least 20 knots (max 33), and have closed circulation.

To become a tropical storm, a further increase in wind speed must occur, as well as rain

becoming organized into rain bands as shown in Fig. 1. The storm also becomes more

concentrated closer to the centre.

Finally, with increasing wind and its circulation and formation of the low pressure eye,

the storm develops into a hurricane (Landsea, 2011).

Hurricanes are classified using a Saffir-Simpson Hurricane Wind Scale (SSHWS), which sorts

them into five categories. The scale ranges from 1 as the lowest wind speed to 5 as the highest

wind speed. Moreover, the storm becomes a hurricane when it reaches 1st category, therefore a

sustainable wind speed of 119km/h or higher has to be recorded, and becomes a major

hurricane if it falls into the 3rd category or higher. SSHWS is often criticized for not including

rainfall density or how fast it moves, ‘only’ the sustainable wind speed. More criteria was

included in the scale until Hurricane Ike in 2007 was cast into a wrong category and caused

damage too strong for the category it was placed in. This scale, however, is only used in two

regions- in the Atlantic Ocean and northern Pacific Ocean, the rest of the areas affected by

tropical cyclones classify their strength by calling them different names, such as typhoons or

cyclones.

Tropical cyclones will rotate clockwise in Southern Hemisphere and counter clockwise in the

Northern Hemisphere due to the influence of Coriolis effect. As they receive their strength and

energy from the evaporation on the surface of the water, they are weakened when on land, due

10 | P a g e

to being cut off from their main power source (NASA, 2012). Hurricanes need warm water to

form, and to sustain its’ ‘life’, as it will cause more evaporation, and therefore form more

humidity in the air. Consequently if a tropical storm forms in a region of unstable stratification it

will start to weaken when moving out of area with high water temperature. It is also observed

that they are unable to form in the immediate vicinity of the Equator, only as far as where the

water reaches the temperature of 26-27 degrees Celsius (Palmen). Tropical cyclones tend to

cause other natural disasters as well, since they often generate tornadoes, floods, high intensity

rain or high waves of water (The Weather Channel: Tropical Storms, 2014). Furthermore, this is

why coastal regions are prone to substantial damage from both high powered winds, as well as

storm surge. Strong wind accompanied by heavy rainfall can, therefore, cause major flooding

inland, such as happened in the case of Hurricane Mitch during which most deaths and property

losses were due to the freshwater flooding. Moreover, as history has shown, considerable

amount of damage and loss of life during a hurricane resulted from powerful wind and wind-

induced storm surges (Emanuel et al., 2006). This is often called a consequence of natural

disaster events.

The worldwide season for tropical cyclones is during the late summer, as water surface

temperatures are at their peak. This, however, varies monthly depending on world regions, as

‘summer’ is during different parts of the year for different continents/countries. More

specifically, Eastern Pacific hurricane season is from May 15th to November 30th, the hurricane

season in the Atlantic basin (this includes Caribbean Sea as well as Gulf of Mexico) runs from

June 1st until November 30th. Pacific Ocean is considered the most prone to formation of tropical

cyclones and most active region being west pacific.

When tropical cyclones form, the National Hurricane Centre names them using an alphabetic

order of six lists which are rotated and recycled every six years. Although, when a hurricane

with a certain name has caused an immense amount of damage or deaths, the name retires from

the list out of respect for the devastation it’s caused.

Monitoring of hurricanes

Although first step in risk analysis of hurricanes is identification and analysis, before that could

happen, data about the threat has to be collected. Therefore, monitoring and observation have

to be carried out, before risk analysis process begins. Monitoring of hurricanes is performed by

several methods, hands-on, such as reconnaissance airplanes, as well as computerized, through

satellites and radars. While monitoring from the distance is the safest option, a more hands-on

approach may present more accurate results.

11 | P a g e

In the United States, the National Weather Service Tropical Prediction Centre (TPC) is in charge

of hurricane prediction, forecasting, tracking and warning of the general public. TPC consists of

three organizations: National Hurricane Center (NHC), Tropical Analysis and Forecast Branch

(TAFB) and Technical Support Branch (TSB) (Fluery, 2011).

Weather satellites

Weather satellites allow for monitoring over great distance, hence giving access to images of

meteorological changes happening on Earth. Satellites are capable of measuring temperatures

inside of a storm, tracking wind speed, detecting snow and rain, cloud depth and movement.

National Oceanic and Atmospheric Association (NOAA), has two types of satellites in the orbit,

each responsible for a different task. Geostational operational environmental satellites (GOES),

hover in the same spot for all their life spans, while polar-orbiting operational environmental

satellites (POES), as the name dictates, circle the Earth around the polar regions 14 times a day

(Oskin, 2013). Although GOES allows for steady inflow of information and data, the imagery

sampled at distance is shot only every 15 minutes (Avila, 2013).

Once a storm starts forming, and small clusters of showers and thunderstorms develop, the job

of a GOES satellite is to stay with the storm from beginning to end of its lifetime and collect all

data. NOAA’s POES is equipped with microwave technology and it passes over the storm twice a

day, at lower altitude, allowing it to collect data all the way down to the ocean’s surface (NOAA,

Hurricane Forecasting).

Radars

Radars job is to measure the severity of the storm; it collects data regarding the rainfall and

allows a look into the inside of the storm. A Doppler radar can measure the estimated rainfall

amount, detects the rain bands, hurricanes eye and the eye wall (The Weather Channel:

Hurricane Tracking & Researching Hurricanes). Radars are usually included in weather

satellites, along with infrared imagery and microwave technology.

Hurricane hunters

Since the 1944, as a more hands-on approach, in the United States, the 53rd Weather

Reconnaissance Squadron, known as hurricane hunters of the U.S. Air Force Reserve fly into the

eye of the hurricane and collect the data from the inside. The planes are equipped with a

Doppler radar, microwave technology and tubes of instruments attached to a parachute called

dropsondes (NOAA, Hurricane Forecasting). Dropsonde is a radio-like instrument which

12 | P a g e

contains tools used to measure humidity, temperature, air pressure and altitude of these

quantities (Fluery, 2011)(The Weather Channel: Hurricane Tracking & Researching

Hurricanes). They also relay the information quickly, as fast as two or four measurements per

second to the nearest aircraft radio (Oskin, 2013). Aircraft reconnaissance aims to fly as low as

safety allows, as most of the information is at the bottom or center of the hurricanes’ cross

section. Determining the exact position of the eye of the hurricane is crucial, and satellite

imagery is not as accurate. Furthermore, hurricane hunters are able to record and transmit the

data regarding wind speed every 30 seconds as they fly through the hurricane (Hurricane

Hunters).

Weather buoys and ships

Weather buoys are scattered across the ocean surface with various measuring devices attached

to them. NOAA has purposely allocated hurricane buoys in storm prone water areas to monitor

the changes in water and air (NOAA, Hurricane Forecasting). Buoys transmit the data by radio

straight to the National Weather Service, sending on-going measurements of wave height, wind

speed and temperature of both water and air. Moreover, majority of ships include weather

gauges and radios, allowing ships to broadcast data to weather centres if ship is in the path of a

storm (Fleury, 2011).

Consequences of hurricanes

Natural disasters are often formed and fuelled by other natural disasters, such as storm surges

being formed under the influence of strong winds of hurricane Katrina. Often times the

consequences of hurricanes are more deadly and devastating than actual primary natural

disaster, such as is this case. The storm surge on the Gulf coast during Katrina was the reason

why tens of thousands of residents were left homeless and without basic supplies.

In places susceptible to hurricanes, buildings are often reinforced to withstand strong winds

and near the shores, houses are built on artificial hills to protect against flood. In New Orleans,

levees, or flood banks were the protective measure against high waves, however, when

Katrina’s enormous storm surge hit the coast, the levees began giving in. The NHC warned in

advance that levees will become overtopped during the storm surge; however, no actual steps

toward strengthening the levees were taken. Before Katrina struck, there were a couple of days

during which certain measures could have been taken to aid in reinforcing the city. Although

now it is known that with waves as high as the ones which hit New Orleans, no amount of flood

banks would suffice.

13 | P a g e

Hurricane Katrina

Hurricanes are among the deadliest and costly natural disasters to roam our planet. Their

forceful nature originates and is fuelled by warm oceanic water basins in the Atlantic and

northern Pacific Ocean. Hurricanes can bring forth incredible destruction to environment,

infrastructures and human life, as well as cause serious health, social, and economic

consequences.

Hurricane Katrina is one of the most infamous natural disasters to occur. Numerous casualties

and countless injuries, along with billions of U.S. dollars in damages make it a very expensive

and overwhelming event. The outcome of the hurricane is the main reason as to why it makes it

interesting to study – has something in risk analysis gone wrong that this hurricane killed this

many people and caused such vast destruction? To find that out, a case study summarizing the

events during hurricane Katrina has to be constructed.

Hurricane Katrina was a 2005 Atlantic hurricane season hurricane, and one of the most

expensive and deadly natural disasters in our history. The death toll reached 1833 fatalities

across five different states in United States, and totalled $125 billion in losses. Tens of

thousands were left homeless and without basic supplies in Mississippi and Louisiana, while the

storm surge flooded countless cities in the coastline. Therefore a proper risk analysis during the

threat is needed to minimize the damages and first of all-protect human life.

The thorough dissection of the events during hurricane Katrina will discover that uncertainty in

regards to frequency and magnitude of the hurricane accredited to the vast damages.

Furthermore, the actions of governors and further persons of power have greatly influenced not

only the actual mitigation process but also the general public and their actions. Lastly, lack of

preparation, due to uncertainty and miss-planned mitigation led to enormous floods and

overpopulated shelters in metropolitan areas.

Katrina was the most costly and disastrous of the several hurricanes occurring in the 2005

Atlantic hurricane season. This hurricane season was the most severe and damaging in recorded

history (Carroll et al., 2005). Katrina affected 92,000 square miles (238,278 square km), caused

1833 fatalities, and left tens of thousands of people homeless and without supplies (Moynihan,

2009). The hurricane ended up affecting Bahamas, Cuba, South Florida, Louisiana, Mississippi,

Alabama, and most of eastern part of North America. Katrina lasted from August 23rd until

14 | P a g e

beginning of September, and managed to make its way from Bahamas to the north of United

States.

On Tuesday the 23rd, a warning was issued by NHC in south of Florida about a tropical

depression stirring in the Bahamas. It formed from remains of a previous tropical depression

which dispersed and later regenerated under influence of a tropical wave, thus creating a more

advanced system. The mentioned above weather system was located 560 km east of Miami

during the warning (Drye, 2005).

Next day, on Wednesday the 24th, the storm has continued to pick up speed and power over

Central Bahamas, upgrading to 11th tropical storm of the season and has been named Katrina. At

this point, the storm’s wind is blowing at 65 km/h and is nearing east of Miami (370 km)(ibid).

Later on in the day, a hurricane warning has been issued for south-eastern coast of Florida

(DeLozier).

Thursday the 25th, in the late afternoon, Katrina has officially become a hurricane of Category 1

with strongest winds at 120 km/h. The forecast stated that the hurricane is headed for Fort

Lauderdale located on the east coast of Florida.

By evening, the eye of the hurricane comes ashore on the south-eastern coast, between

Hallandale Beach and North Miami Beach; the top winds have increased to 130 km/h and

Katrina is now slowly moving though Florida (Drye, 2005). Three men were killed by different

incidents including trees (Carroll et al., 2005), and at least five people died from carbon

monoxide poisoning, while one was killed during debris clean up. Overall, fourteen people were

killed in Florida during hurricane Katrina (CDC, 2006).

According to reports, Katrina changed its direction and instead of heading across the peninsula,

the hurricane pushed on southwest. During the night, Florida Power and Light announced that

over a million households in three counties were left without electricity (ibid)(Ross, 2007). The

governor of Florida, Jeb Bush declares a State of Emergency in Florida (DeLozier), and calls a

press conference during which he warns people to take the hurricane seriously regardless

which Category it may fall into. Evacuation was urged toward public residing in mobile homes,

barrier islands or low-lying areas, however, it is mostly voluntary. Mandatory evacuation was

only mandatory around Boca Raton, in the southern Palm Beach County. News reporters,

however, did encounter people using plywood to cover windows in hopes of protection against

powerful wind (Carroll et al., 2005).

15 | P a g e

Furthermore, the rain caused flooding across southern Florida, a 1.64m high storm surge in

Pensacola (Knabb et al., 2005), along with a tornado in Florida Keys. Overall, the damage

estimated after Katrina in South Florida added up to $623 U.S. million.

On Friday, August 26th, at 1 a.m. while still on land, Katrina weakened and became reclassified

as a tropical storm. However, a couple of hours later, as soon as Katrina crossed out of Florida

and into the warm waters of Gulf of Mexico, the weather system has immediately began picking

up strength and turned back into a 1st category hurricane (Drye, 2005). The winds have

increased to 160 km/h and Katrina has upgraded to a 2nd category hurricane. In the early

afternoon, the National Hurricane Centre announces that the predicted forecast for Katrina may

be wrong and instead of staying in Florida, it will in fact head over towards Mississippi and

Louisiana (Moynihan, 2009). Governor of Louisiana, Kathleen Blanco and governor of

Mississippi, Haley Barbour, declare the State of Emergency in their respective states (Ross,

2007), therefore Federal Emergency Management Agency (FEMA) is deployed and on standby

(DeLozier). White House declares impeding disaster area, and 10,000 National Guard troops

from Pentagon are sent along the Gulf of Mexico (ThinkProgress, 2005)(DeLozier).

At 5 a.m. on Saturday the 27th, hurricane Katrina has been named a category 3 hurricane, with

winds reaching 185 km/h (Drye, 2005). A hurricane warning was issued for northern Gulf Coast

and for Louisiana’s south-eastern coast (DeLozier), and several cities and towns along the coast

have been ordered mandatory evacuation (Associated Press, 2005). During the day a hurricane

hunter aircraft flew into Katrina to take measurements, and according to meteorologist Chris

Landsea, the ‘circulation (of the hurricane) covered the entire Gulf of Mexico (Drye, 2005).

Moreover, at the request of Kathleen Blanco, President George Bush declares a Federal State of

Emergency for Louisiana, therefore giving FEMA and Department of Homeland Security (DHS)

authority to provide aid (Ross, 2007)(ThinkProgress, 2005). In the afternoon, National

Hurricane Director Max Mayfield advises New Orleans Mayor Ray Nagin to induce mandatory

evacuation. Ray Nagin, however, declares State of Emergency and calls for voluntary evacuation,

while in Mississippi, the governor orders mandatory evacuation for Hancock County (DeLozier).

Residents of the coastal gulf rush to evacuate and cause large traffic jams on major streets and

gas stations (Associated Press, 2005).

Just after midnight on the next day, Sunday the 28th, hurricane strengthens into a category 4

hurricane with winds reaching 235 km/h, and it is slowly reaching the Mississippi River (Drye,

2005). By 7 a.m., Katrina’s sustained winds increase to 280 km/h, however, gusts of wind with

speed of 310 km/h have been recorded as well. At this time the hurricane becomes classified as

a category 5 hurricane (DeLozier) and with winds this strong it is already known to be one of

16 | P a g e

the strongest hurricanes to ever occur in the Atlantic (Drye, 2005). General public is warned

that the flood bank on the coastline may give into Katrina, and consequences may be severe

(ThinkProgress, 2005).

At 9:30 a.m., Ray Nagin orders a mandatory evacuation and tens of thousands of people are

rushing out of the city (Drye, 2005). The Mayor also announced that Regional Transit Buses will

be available at 12 locations throughout the city to pick residents up and take them to places of

refugee (DeLozier). A Superdome has been opened in Louisiana, largest out of 10 shelters

available or ‘refuges of last resort’, for people who were unable to leave the city

(ibid)(Associated Press, 2005). Evacuation orders are being posted all across the coast, and

numerous people evacuate, while 10,000 took shelter in the Superdome, along with protection

of National Guard. Governor of Alabama, Bob Riley declares the State of Emergency, whereas

FEMA supplies water, food and essential goods to Georgia and Texas in preparation of what’s to

come (DeLozier). President Bush declares State of Emergency for Mississippi and Alabama, and

declares Florida as a federal disaster area (Ross, 2007). In the late afternoon, the NHC warns

that levees in New Orleans are becoming overtopped and significant storm surge will cause

major flooding along the central and north-eastern coast of Gulf of Mexico (Drye, 2005).

On early morning on Monday the 29th, Katrina is about to come ashore Louisiana with strongest

winds at about 235 km/h, its eye 115 km from New Orleans. Weather buoys stationed 80 km

east from the river’s mouth records waves as high as 12 meters (ibid). Later that morning levees

in New Orleans are breached, power is lost due to partial flooding of the city due to water

pouring through street canals (Ross, 2007)(Drye, 2005). At 11 a.m., the eye of the hurricane

comes ashore again near the border of Louisiana and Mississippi, Katrina’s strongest winds and

storm surge slam into Mississippi’s coast causing devastating damage right away (Drye, 2005).

Although hurricane’s winds are gradually declining in strength, and its strongest winds have

weakened to 170 km/h and declining (ibid), the winds have managed to damage the roof of the

Superdome. Airports have closed in New Orleans, Baton Rouge, Biloxi, Mobile and Pensacola

(Associated Press, 2005), and gulf coast refineries were vastly damaged and had to be closed for

the time being (DeLozier). At this point, 3,500 National Guardsmen assist 1,500 police officers in

rescue operations (Ross, 2007), whilst Coast Guard managed to rescue 1,200 from the flood

(DeLozier). At the end of the day, Terry Ebbert, New Orleans’s director of Homeland Security

made a statement that everyone who didn’t manage to get away from the way of the storm, ‘it

was their last night on this Earth’ (Ross, 2007).

By Tuesday, 80% of New Orleans was flooded, and general conditions in Superdome and other

shelters began deteriorating quickly. With law enforcement occupied, out of desperation for

17 | P a g e

lack of food and water, general public resorted to looting and burglary (Ross, 2007). Situation at

the Superdome was getting out of hand; people were forced to urinate on the floor because of

the lack of sanitation, whilst at least two people, including a child were sexually assaulted

(ThinkProgress, 2005). Katrina continued to lose wind speed and became less and less

threatening as days went by, until it finally downgraded to a tropical depression (Drye, 2005).

An estimated number of 50,000 to 100,000 people still remained in New Orleans, in the

Superdome, on tall buildings roofs, and various other places which under circumstances quickly

were adapted into becoming shelters (DeLozier). In the upcoming days, Blanco and Nagin order

a total evacuation, and slowly people are leaving the crisis area, by boats, helicopters and buses.

Governors desperately ask for assistance and Congress finally approves for $10,5 billion dollars

in aid for immediate rescue and relief for Hurricane Katrina (Ross, 2007). On Friday, September

2nd, U.S. National Guard troops arrive in New Orleans with supply trucks loaded with food and

water for the people still stranded in the city (Drye, 2005). A week later President Bush,

unsatisfied with efforts made by government, requests another $52 billion in aid, and is

approved by Congress (DeLozier).

In conclusion, the final insured damage after Katrina rounded up to $40 to 60 billion (Risk

analysis Solutions Inc., 2005) or overall $125 billion in losses (Emanuel at al., 2006), while death

toll has risen to 1833 people (Moynihan, 2009)

18 | P a g e

Discussion

Several factors during hurricane Katrina may be contributing to the outcome and aftermath of

this natural disaster. Hurricanes are one of the most destructive natural forces on our planet,

and Katrina was the evidence of that. The overall death toll and caused damage is, however, the

proof that risk analysis is not a perfect system after all.

Monitoring and observation are the step prior to risk analysis, as they are what sets the entire

process in motion. The unpredictability of hurricanes is what makes them so destructive and

terrifying, therefore the need to detect them as quickly as possible is crucial. The fact that

weather systems are only truly discovered after they form makes them that much more

daunting, therefore monitoring is vital to determine whether the system will strengthen into an

actual hurricane or disperse into something non-threatening. According to Miller, this inability

to predict the outcome is to be attributed to uncertainty. Uncertainty stands for limited amount

of knowledge and data in hurricane risk modelling and management in hurricanes. Having

limited data on previous hurricanes, such as inside measurements taken by hurricane hunters

can be a colossal factor in risk analysis.

Miller states that uncertainty could potentially exist due to incorrect choice of measurements

and parameters within the risk, combined with the finite number of historical hurricane

observations. This, however, only addresses the uncertainty dealing with frequency, and there

is also uncertainty which corresponds to severity. During hurricane Katrina, the strength of the

hurricane changed drastically throughout the week the weather system was active. Although it

was known that warm water basins fuel the hurricane, the uncertainty of whether the hurricane

would head over the Gulf of Mexico or disperse over land, may have been the factor which in the

end cost lives.

The NHC uses the cone of uncertainty when officially issuing the track forecast for a hurricane,

which is used to estimate the uncertainty. The projected forecast works by showing the current

location of the weather system and the predicted path, however, as the cone moves further from

the current location of the hurricane, the uncertainty increases and cone becomes wider. During

Katrina, the uncertainty was as usual very high, however, as the path of the hurricane diverted

in a completely different direction, the areas which were assumed to be under the influence

were not, and new territory was in the path. The uncertainty and error in forecasting left little

to no time for mitigation and might have been the cause of loss of credibility on the part of NHC

19 | P a g e

and persons of power. This level of uncertainty could definitely have been the factor which

influenced the outcome of the hurricane Katrina.

The NHC is the branch responsible for issuing the official intensity and track forecasts for

hurricanes in the Atlantic and North Eastern Pacific regions (Quiring et al., 2013). NHC

configures models basing their knowledge on incoming data from satellites, radars and

foremost the hurricane hunter aircrafts. Combing this newly composed data with the data

collection from previously occurring hurricanes allows them to craft a model which will then be

presented to the general public. However, NHC is aware that although years of research and

model development have led to reduction in track forecast errors, the hurricane forecasts are

still nowhere near perfect (ibid). This can cause an enormous problem for not only general

public but also for people who depend on these forecast models to further the risk analysis

process.

During risk analysis, forecasts aid in determining the further mode of actions when it comes to

hurricane mitigation. Using drafted forecasts, people of power can design a proper mitigation

plan to suit the affected areas and hopefully lessen the damage of impact. However, if a forecast

is incorrect and the hurricane changes paths and diverts toward another location, that location

can undergo vast damages due to inability to prepare for the incoming natural disaster

(Emanuel et al., 2006). In case of Katrina, the forecasting was also slightly misleading, as in the

original pathway model, Katrina was meant to head across Florida, rather than divert toward

Louisiana and Mississippi. Luckily, the fact that Katrina had to cross the Gulf of Mexico first,

gave another day for preparations for the two states before hitting them with enormous force.

Although one day is not much time, in this case it might have been the factor which saved

thousands of lives, due to earlier evacuation. Therefore, further research in model development

is crucial to reducing the error margin in hurricane forecasting and consequently, the risk

analysis.

Forecasting and risk models are never 100% accurate, therefore a slight change in hurricane

pathway can lead to devastating results in terms of damages and credibility of people of power

who announced the hurricane warning. Loss of credibility could occur if general public is

expecting a hurricane in their area, but the forecast is misleading and the hurricane never

befalls, therefore peoples’ evacuation and preparation would be aimless. People living in

hurricane prone areas might also be less inclined to participate in evacuation or mitigation

plans, as they are exposed to numerous hurricanes, therefore building up tolerance. Not all

hurricanes are high category and life threatening, and living in areas where hurricanes frequent

can make people tired of constant evacuation.

20 | P a g e

The role of general public in risk analysis is of importance, as the outcome of the hurricane is

first-handily tied to citizens whom were affected. The outcome of mitigation and overall risk

analysis is done to minimize the damage to buildings, environment, and foremost to protect

human life. People of power are solely in charge of mitigation in their respective counties and

states in United States; however, it is the public’s willingness and cooperation that fuels the

process.

A problem during mitigation might arise when a resident is unable to leave the area designated

for evacuation due to health or private, perhaps economic problems. According to FoxNews,

71% of the victims of Katrina and Louisiana were older than sixty, while 47% of those were

above seventy-five (2006). It has also been published that at least sixty-eight of the elderly were

found in nursing homes which were abandoned by the caretakers when the hurricane struck

(ibid). The elderly were left without care, power and possibility of evacuation, therefore leaving

them stranded in the disaster area. As most of the deaths claimed by Katrina in Louisiana were

elderly, it may also mean that several of them were unable to take care of themselves, therefore

unable to drive and consequently evacuate with other residents. Moreover, leaving people who

were incapable of evacuating behind could possibly be the factor which explains the high death

toll in Louisiana.

Furthermore, as explained in previous subsection, people often tend to ignore or disregard the

orders given by authority, even in times of crisis. Although during hurricane Katrina, governor

of Florida, Jeb Bush warned the general public to stay inside and to take natural disasters

seriously, no regard to the strength, people still did get killed. A man in Fort Lauderdale was

killed when a ficus tree fell on his car and brought power lines down with it. In Plantation,

another man died when going outside to inspect damage to his mobile home and got crushed by

a tree. Third deadly incident in Florida was in Cooper City, where a man accidently crashed his

car into a tree while avoiding a tree on a road (Carroll et al., 2005). All of these incidents were of

people who left their homes and disregarded the orders to stay inside and wait the hurricane

out. Free will, however, cannot be manipulated and people can be under mandatory evacuation

and refusing to listen.

During mitigation, the governor of Louisiana was advised by National Hurricane Director Max

Mayfield to encourage mandatory evacuation. The governor weighed his value judgments and

chose to not listen and call upon voluntary evacuation instead. The reason might have been to

lessen the panic and traffic, although the result was nearly identical. Inducing mandatory

evacuation early on may have cause majority of the residents to leave the city, therefore not

leaving as many behind without sufficient supplies.

21 | P a g e

Conclusion

Risk analysis in natural disasters is a lengthy process, however, in case of hurricanes; the

process requires the data collection and decisions to be made rather quickly. The case study of

hurricane Katrina presents several various occurrences during the progression of the hurricane

which influence the outcome. Each of these details led to the unravelling of events that fuelled

the risk analysis and therefore, the consequences of the aftermath.

The outcome of the hurricane Katrina was terribly devastating, and the choices made by

authorities and generally persons of power impacted every step of the process. The error in

hurricane pathway forecasting also largely contributed to the outcome, as less time for

preparations was left in the coast of the Gulf of Mexico. This, however, cannot be blamed on the

scientists or persons of power, as a hurricane is unpredictable and a perfect hurricane pathway

model is yet to exist. The uncertainty in hurricane pathway, strength and overall hurricane

frequency is something that has to be taken into consideration, as this is where the margin of

error is at large. Uncertainty in forecasting was the major downfall of the risk analysis process,

as this is where all the important decisions were made. Wrong pathway forecast may have been

the factor which also lessened the credibility of the authorities and left little time for measures

to be taken.

Yet another factor which highly influenced the devastating outcome of the hurricane was the

consequences which were formed due to Katrina. The uncertainty dealing with probability and

severity of consequences was underestimated and therefore no preparation was made for the

approaching storm surge. The storm surge which slowly flooded the coast was massive, and

evacuation should have been induced as quickly as possible. Although normally reinforcing the

levees may have been a factor which saved the coast, in this particular case, the surge was too

high and powerful.

Furthermore, the risk analysis process may have been sped up and would not leave time for

authorities to thoroughly weigh their options in regards to actions they took. Although having

more time would be ideal, in cause of hurricanes it’s a rare occurring commodity. Lastly, to

minimize the damages in hurricanes, understanding of the threat, and continuous improving of

technology and modelling systems is the future of risk analysis.

22 | P a g e

References

Associated Press. (2005). Fast Facts: Hurricane Katrina Timeline. FoxNews. Web Accessed May

8, 2014 < http://www.foxnews.com/story/2005/09/03/fast-facts-hurricane-katrina-

timeline/>

Avila, Leanne. (2013). Hurricane Monitoring in 2013: Improving the Tools and Our

Understanding. News Articles. Space Science and Engineering Center. Web Accessed

May 6, 2014 < https://www.ssec.wisc.edu/news/articles/3030>

Carroll, Jason., Jeras, Jacqui., Marciano, Rob., Zarrella, Ed. (2005). Katrina hit Florida: 3 dead; 1

million dark. Weather. CNN. Web Accessed May 8, 2014 <

http://edition.cnn.com/2005/WEATHER/08/25/tropical.storm/>

CDC (2006). Morbidity and Mortality Weekly Report. Centers for Disease Control and

Prevention. Web Accessed April 27, 2014

<http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5509a5.htm>

DeLozier. Hurricane Katrina Timeline. The Brookings Institution. Web Accessed May 6, 2014 <

http://www.brookings.edu/fp/projects/homeland/katrinatimeline.pdf>

Drye, Willie. (2005). Hurricane Katrina: The Essential Time Line. National Geographic News.

Web Accessed May 7, 2014 <

http://news.nationalgeographic.com/news/2005/09/0914_050914_katrina_timeline.ht

ml>

Emanuel, Kerry., Ravela, Sai., Viviant, Emmanuel., Risi, Camille. (2006). A Statistical

Deterministic Approach to Hurricane Risk Assessment. American Meterorological

Society, Volume 87, Issue 3. March 2006. Pages 299-300

Fluery, Maureen, K. (2011). How Hurricanes are Forecasted and Monitored. Web Accessed, May

2, 2014 < https://suite.io/maureen-k-fleury/59wt233>

FoxNews. (2006). FOX Facts: Hurricane Katrina Damage. Fox News. Web Accessed May 10, 2014

< http://www.foxnews.com/story/2006/08/29/fox-facts-hurricane-katrina-damage/>

23 | P a g e

Hungr U., Fell, R., Couture R., Eberhardt E. (2005). A Framework for Landslide Risk Assessment

and Management. Landslide Risk analysis. Proceedings of the International Conference

on Risk analysis. Vancouver Canada

Hurricane Hunters. FAQ. Hurricane Hunters Association. Supporting the Air Force Reserve 53rd

Weather Reconnaissance Squadron ‘Hurricane Hunters’. Web Accessed May 3, 2014

<http://www.hurricanehunters.com/faq.htm>

Knabb, Richard., Rhome, Jamie & Brown, Daniel. (2005). Tropical Cyclone Report. National

Hurricane Center. NOAA. Web Accessed May 9, 2014 <

http://www.nhc.noaa.gov/pdf/TCR-AL122005_Katrina.pdf>

Landsea, Chris. (2011). What is a tropical disturbance, a tropical depression or a tropical storm?

FAQ. Hurricane Research Division: Atlantic Oceanographic & Meteorological Laboratory.

Web Accessed April 26, 2014 < http://www.aoml.noaa.gov/hrd/tcfaq/A5.html>

Miller, David. Uncertainty in Hurricane Risk Modeling and Implications for Securitization. Guy

Carpenter & Company.

Moynihan, Donald, P. (2009). The Response to Hurricane Katrina. Risk Governance Deficits: An

analysis and illustration of the most common deficits in risk governance. Web Accessed

May 7, 2014 <http://irgc.org/wp-

content/uploads/2012/04/Hurricane_Katrina_full_case_study_web.pdf>

NASA (National Aeronautics and Space Administration). (2012). About Hurricanes. Web

Accessed April 26, 2014 < http://www.nasa.gov/mission_pages/hurricanes/features/>

NOAA Hurricane Forecasting (National Oceanic & Atmospheric Administration). United States

Department of Commerce. Hurricane Forecasting Fact Sheet. Web Accessed May 3, 2014

< http://www.nws.noaa.gov/pa/files/hurricane%20forecasting.pdf>

Oskin, Becky. (2013). Incredible Technology: How to Track Hurricanes. Live Science. Web

Accessed May 2, 2014 < http://www.livescience.com/39678-incredible-tech-track-

hurricanes.html>

Palmen, E. On the Formation and Structure of Tropical Hurricanes. University of Chicago. Pages

Quiring, Steven M., Schumacher, Andrea B., Guikema, Seth D. (2013). Incorporating Hurricane

Forecast Uncertainty into a Decision Support Application for Power Outage Modeling.

American Meteorological Society: Bulletin of the American Meteorological Society. Pages

1-7

24 | P a g e

Ready. Risk Assessment. Business Planning. Ready: Prepare, Plan, Stay Informed. Web Accessed

April 10, 2014 < http://www.ready.gov/risk-assessment>

Risk analysis Solutions Inc. (2005). Hurricane Katrina: Profile of a Super Cat, Lessons and

Implications for Catastrophe Risk analysis. Web Accessed 14 March, 2014

<https://support.rms.com/publications/katrinareport_lessonsandimplications.pdf>

Ross, Shmuel. (2007). Hurricane Katrina Timeline. Infoplease. Web Accessed May 6, 2014 <

http://www.infoplease.com/spot/hurricanekatrinatimeline.html>

The Weather Channel: Hurricane Tracking & Researching Hurricanes. The Weather Channel.

Web Accessed May 2, 2014 <http://www.weather.com/outlook/weather-

news/hurricanes/articles/hurricane-tracking-hurricanes_2010-05-24>

The Weather Channel: Tropical Storms. (2014). Tropical Storms. The Weather Channel. Web

Accessed April 13, 2014 < http://www.aoml.noaa.gov/hrd/tcfaq/A5.html>

ThinkProgress. (2005). Katrina Timeline. ThinkProgress. Web Accessed May 5, 2014 <

http://thinkprogress.org/report/katrina-timeline/>

Watson, John T., Gayer, Michelle., Connolly, Marie A. (2007). Epidemics after Natural Disasters.

Emerging Infectious Diseases, Volume 13, No. 1. Pages 1-2

Wolchover, Natalie. (2011). Why Is the Eye of a Hurricane Calm? Life’s Little Mysteries. Live

Science. Web Accessed April 12, 2014 <http://www.livescience.com/15805-calm-

hurricane-eye.html>